R version 4.2.3 (2023-03-15)
Platform: x86_64-apple-darwin17.0 (64-bit)
Running under: macOS Big Sur ... 10.16
Matrix products: default
BLAS: /Library/Frameworks/R.framework/Versions/4.2/Resources/lib/libRblas.0.dylib
LAPACK: /Library/Frameworks/R.framework/Versions/4.2/Resources/lib/libRlapack.dylib
locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
attached base packages:
[1] stats graphics grDevices utils datasets methods base
loaded via a namespace (and not attached):
[1] htmlwidgets_1.6.2 compiler_4.2.3 fastmap_1.1.1 cli_3.6.1
[5] tools_4.2.3 htmltools_0.5.5 rstudioapi_0.14 yaml_2.3.7
[9] rmarkdown_2.23 knitr_1.43 jsonlite_1.8.7 xfun_0.39
[13] digest_0.6.32 rlang_1.1.1 evaluate_0.21
Load necessary libraries (you can add more as needed).
library(arrow)
Warning:
It appears that you are running R and Arrow in emulation (i.e. you're
running an Intel version of R on a non-Intel mac). This configuration is
not supported by arrow, you should install a native (arm64) build of R
and use arrow with that. See https://cran.r-project.org/bin/macosx/
Attaching package: 'arrow'
The following object is masked from 'package:utils':
timestamp
library(memuse)library(pryr)library(R.utils)
Loading required package: R.oo
Loading required package: R.methodsS3
R.methodsS3 v1.8.2 (2022-06-13 22:00:14 UTC) successfully loaded. See ?R.methodsS3 for help.
R.oo v1.25.0 (2022-06-12 02:20:02 UTC) successfully loaded. See ?R.oo for help.
Attaching package: 'R.oo'
The following object is masked from 'package:R.methodsS3':
throw
The following objects are masked from 'package:methods':
getClasses, getMethods
The following objects are masked from 'package:base':
attach, detach, load, save
R.utils v2.12.2 (2022-11-11 22:00:03 UTC) successfully loaded. See ?R.utils for help.
Attaching package: 'R.utils'
The following object is masked from 'package:arrow':
timestamp
The following object is masked from 'package:utils':
timestamp
The following objects are masked from 'package:base':
cat, commandArgs, getOption, isOpen, nullfile, parse, warnings
── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
✖ purrr::compose() masks pryr::compose()
✖ lubridate::duration() masks arrow::duration()
✖ tidyr::extract() masks R.utils::extract()
✖ dplyr::filter() masks stats::filter()
✖ dplyr::lag() masks stats::lag()
✖ purrr::partial() masks pryr::partial()
✖ dplyr::where() masks pryr::where()
ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors
library(plotly)
Attaching package: 'plotly'
The following object is masked from 'package:ggplot2':
last_plot
The following object is masked from 'package:arrow':
schema
The following object is masked from 'package:stats':
filter
The following object is masked from 'package:graphics':
layout
Display your machine memory.
memuse::Sys.meminfo()
Totalram: 8.000 GiB
Freeram: 1.488 GiB
In this exercise, we use tidyverse (ggplot2, dplyr, etc) to explore the MIMIC-IV data introduced in homework 1 and to build a cohort of ICU stays.
Q1. Visualizing patient trajectory
Visualizing a patient’s encounters in a health care system is a common task in clinical data analysis. In this question, we will visualize a patient’s ADT (admission-discharge-transfer) history and ICU vitals in the MIMIC-IV data.
Q1.1 ADT history
A patient’s ADT history records the time of admission, discharge, and transfer in the hospital. This figure shows the ADT history of the patient with subject_id 10001217 in the MIMIC-IV data. The x-axis is the calendar time, and the y-axis is the type of event (ADT, lab, procedure). The color of the line segment represents the care unit. The size of the line segment represents whether the care unit is an ICU/CCU. The crosses represent lab events, and the shape of the dots represents the type of procedure. The title of the figure shows the patient’s demographic information and the subtitle shows top 3 diagnoses.
Do a similar visualization for the patient with subject_id 10013310 using ggplot.
Hint: We need to pull information from data files patients.csv.gz, admissions.csv.gz, transfers.csv.gz, labevents.csv.gz, procedures_icd.csv.gz, diagnoses_icd.csv.gz, d_icd_procedures.csv.gz, and d_icd_diagnoses.csv.gz. For the big file labevents.csv.gz, use the Parquet format you generated in Homework 2. For reproducibility, make the Parquet folder labevents_pq available at the current working directory hw3, for example, by a symbolic link. Make your code reproducible.
Rows: 299712 Columns: 6
── Column specification ────────────────────────────────────────────────────────
Delimiter: ","
chr (2): gender, anchor_year_group
dbl (3): subject_id, anchor_age, anchor_year
date (1): dod
ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
Rows: 431231 Columns: 16
── Column specification ────────────────────────────────────────────────────────
Delimiter: ","
chr (8): admission_type, admit_provider_id, admission_location, discharge_l...
dbl (3): subject_id, hadm_id, hospital_expire_flag
dttm (5): admittime, dischtime, deathtime, edregtime, edouttime
ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
Rows: 1890972 Columns: 7
── Column specification ────────────────────────────────────────────────────────
Delimiter: ","
chr (2): eventtype, careunit
dbl (3): subject_id, hadm_id, transfer_id
dttm (2): intime, outtime
ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
Rows: 669186 Columns: 6
── Column specification ────────────────────────────────────────────────────────
Delimiter: ","
chr (1): icd_code
dbl (4): subject_id, hadm_id, seq_num, icd_version
date (1): chartdate
ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
Rows: 4756326 Columns: 5
── Column specification ────────────────────────────────────────────────────────
Delimiter: ","
chr (1): icd_code
dbl (4): subject_id, hadm_id, seq_num, icd_version
ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
Rows: 85257 Columns: 3
── Column specification ────────────────────────────────────────────────────────
Delimiter: ","
chr (2): icd_code, long_title
dbl (1): icd_version
ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
Rows: 109775 Columns: 3
── Column specification ────────────────────────────────────────────────────────
Delimiter: ","
chr (2): icd_code, long_title
dbl (1): icd_version
ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
Rows: 299712 Columns: 6
── Column specification ────────────────────────────────────────────────────────
Delimiter: ","
chr (2): gender, anchor_year_group
dbl (3): subject_id, anchor_age, anchor_year
date (1): dod
ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
Filter data for the patient with subject_id 10013310
# Make sure the data is in right formatpatient_procedures$chartdate <-as.POSIXct(patient_procedures$chartdate)patient_transfers <- patient_transfers[complete.cases(patient_transfers$careunit, patient_transfers$intime, patient_transfers$outtime), ]patient_procedures$long_title <-as.factor(patient_procedures$long_title)
# Let the y-axis be the type of event (ADT, lab, procedure)convert_to_text <-function(x) { y_labels <-c(" ","Procedure", " "," Lab", " ", "ADT") y_labels[ceiling(x *length(y_labels))]}# Make the legend more readabletruncate_legend <-function(x, max_chars) {ifelse(nchar(x) > max_chars, substr(x, 1, max_chars), x)}# Find the top three diagnosestitle_freq <-table(patient_diagnoses$long_title)top_three_titles <-names(sort(title_freq, decreasing =TRUE)[1:3])subtitle <-sprintf("\n%s\n%s\n%s", top_three_titles[1], top_three_titles[2], top_three_titles[3])
ICU stays are a subset of ADT history. This figure shows the vitals of the patient 10001217 during ICU stays. The x-axis is the calendar time, and the y-axis is the value of the vital. The color of the line represents the type of vital. The facet grid shows the abbreviation of the vital and the stay ID.
Do a similar visualization for the patient 10013310.
Rows: 73181 Columns: 8
── Column specification ────────────────────────────────────────────────────────
Delimiter: ","
chr (2): first_careunit, last_careunit
dbl (4): subject_id, hadm_id, stay_id, los
dttm (2): intime, outtime
ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
d_items <-read_csv("~/mimic/icu/d_items.csv.gz")
Rows: 4014 Columns: 9
── Column specification ────────────────────────────────────────────────────────
Delimiter: ","
chr (6): label, abbreviation, linksto, category, unitname, param_type
dbl (3): itemid, lownormalvalue, highnormalvalue
ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
# Select the patienticu_stays_patient <- icu_stays %>%filter(subject_id ==10013310)chartevents_patient <- chartevents %>%filter(subject_id ==10013310)chartevents_patient <-as_tibble(chartevents_patient)# Select the items that we are interested incondition <- d_items$linksto =="chartevents"& d_items$abbreviation %in%c("HR", "NBPd", "NBPs", "RR", "Temperature F")filtered_items <- d_items$itemid[condition]d_items_need <- d_items %>%filter(itemid %in% filtered_items)
# Merge the datachartevents_patient <- chartevents_patient %>%filter(itemid %in% filtered_items) %>%left_join(d_items_need, by ="itemid") %>%select(subject_id, stay_id,charttime, value, valuenum, abbreviation, itemid)
# Make sure the data is in right formatchartevents_patient$charttime <-as.POSIXct(chartevents_patient$charttime)chartevents_patient$stay_id <-as.factor(chartevents_patient$stay_id)chartevents_patient$abbreviation <-as.factor(chartevents_patient$abbreviation)
subject_id,hadm_id,stay_id,first_careunit,last_careunit,intime,outtime,los
10000032,29079034,39553978,Medical Intensive Care Unit (MICU),Medical Intensive Care Unit (MICU),2180-07-23 14:00:00,2180-07-23 23:50:47,0.4102662037037037
10000980,26913865,39765666,Medical Intensive Care Unit (MICU),Medical Intensive Care Unit (MICU),2189-06-27 08:42:00,2189-06-27 20:38:27,0.4975347222222222
10001217,24597018,37067082,Surgical Intensive Care Unit (SICU),Surgical Intensive Care Unit (SICU),2157-11-20 19:18:02,2157-11-21 22:08:00,1.1180324074074075
10001217,27703517,34592300,Surgical Intensive Care Unit (SICU),Surgical Intensive Care Unit (SICU),2157-12-19 15:42:24,2157-12-20 14:27:41,0.9481134259259258
10001725,25563031,31205490,Medical/Surgical Intensive Care Unit (MICU/SICU),Medical/Surgical Intensive Care Unit (MICU/SICU),2110-04-11 15:52:22,2110-04-12 23:59:56,1.338587962962963
10001884,26184834,37510196,Medical Intensive Care Unit (MICU),Medical Intensive Care Unit (MICU),2131-01-11 04:20:05,2131-01-20 08:27:30,9.171817129629629
10002013,23581541,39060235,Cardiac Vascular Intensive Care Unit (CVICU),Cardiac Vascular Intensive Care Unit (CVICU),2160-05-18 10:00:53,2160-05-19 17:33:33,1.3143518518518518
10002155,20345487,32358465,Medical Intensive Care Unit (MICU),Medical Intensive Care Unit (MICU),2131-03-09 21:33:00,2131-03-10 18:09:21,0.8585763888888889
10002155,23822395,33685454,Coronary Care Unit (CCU),Coronary Care Unit (CCU),2129-08-04 12:45:00,2129-08-10 17:02:38,6.178912037037037
Rows: 73181 Columns: 8
── Column specification ────────────────────────────────────────────────────────
Delimiter: ","
chr (2): first_careunit, last_careunit
dbl (4): subject_id, hadm_id, stay_id, los
dttm (2): intime, outtime
ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
Q2.2 Summary and visualization
How many unique subject_id? Can a subject_id have multiple ICU stays? Summarize the number of ICU stays per subject_id by graphs.
Rows: 431231 Columns: 16
── Column specification ────────────────────────────────────────────────────────
Delimiter: ","
chr (8): admission_type, admit_provider_id, admission_location, discharge_l...
dbl (3): subject_id, hadm_id, hospital_expire_flag
dttm (5): admittime, dischtime, deathtime, edregtime, edouttime
ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
Q3.2 Summary and visualization
Summarize the following information by graphics and explain any patterns you see.
number of admissions per patient
admission hour (anything unusual?)
admission minute (anything unusual?)
length of hospital stay (from admission to discharge) (anything unusual?)
All dates in the database have been shifted to protect patient confidentiality. Dates will be internally consistent for the same patient, but randomly distributed in the future. Dates of birth which occur in the present time are not true dates of birth. Furthermore, dates of birth which occur before the year 1900 occur if the patient is older than 89. In these cases, the patient’s age at their first admission has been fixed to 300.
We can see that over 100,000 patients have only one admission, about 35,000 patients have two admissions, and about 15,000 patients have three admissions. Only 15.43% of patients have more than 3 admissions.
Admission hour:
admissions_tble$admittime <-as.POSIXct(admissions_tble$admittime, format ="%Y-%m-%d %H:%M:%S")admissions_tble$hour <-hour(admissions_tble$admittime)hour_counts <- admissions_tble %>%count(hour)
plot <-ggplot(hour_counts, aes(x = hour, y = n)) +geom_bar(stat ="identity") +labs(title ="Count of Admissions by Hour",x ="Hour",y ="Count") +theme_minimal()plotly_plot <-ggplotly(plot)plotly_plot
From the graph, we can see that the number of admissions is relatively low between 1:00 ~ 6:00, and 8:00 ~ 13:00, and relatively high between 16:00 ~ 24:00. It implies the fact that most ICU admissions occur during the night and afternoon.
plot <-ggplot(minute_counts, aes(x = minute, y = n)) +geom_bar(stat ="identity") +labs(title ="Count of Admissions by Minute",x ="Minute",y ="Count") +theme_minimal()plotly_plot <-ggplotly(plot)plotly_plot
Most of the admissions occur at every quarter of the hour, and have a relative avergae occurance at the rest of the minutes.
Length of hospital stay:
admissions_tble$dischtime <-as.POSIXct(admissions_tble$dischtime, format ="%Y-%m-%d %H:%M:%S")admissions_tble$admittime <-as.POSIXct(admissions_tble$admittime, format ="%Y-%m-%d %H:%M:%S")admissions_tble$los <-as.numeric(difftime(admissions_tble$dischtime, admissions_tble$admittime, units ="days"))
plot <-ggplot(data = admissions_tble) +geom_histogram(mapping =aes(x = los),bins =100) +labs(title ="Distribution of Length of Hospital Stay",x ="Length of Hospital Stay",y ="Count") +theme_minimal()plotly_plot <-ggplotly(plot)plotly_plot
From the graph, we can see that the distribution of length of hospital stay is right-skewed, with most patients stay in the hospital for less than 10 days, and it’s unusual that some patients has a length of hospital stay that is negative number.
Rows: 299712 Columns: 6
── Column specification ────────────────────────────────────────────────────────
Delimiter: ","
chr (2): gender, anchor_year_group
dbl (3): subject_id, anchor_age, anchor_year
date (1): dod
ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
Q4.2 Summary and visualization
Summarize variables gender and anchor_age by graphics, and explain any patterns you see.
gender_counts <- patients_tble %>%count(gender)ggplot(gender_counts, aes(x ="", y = n, fill = gender)) +geom_bar(stat ="identity", width =1) +coord_polar("y", start =0) +labs(title ="Gender Distribution",fill ="Gender") +scale_fill_manual(values =c("lightpink1", "lightblue2")) +geom_text(aes(label =paste0(n, " (", round((n/sum(n))*100), "%)")), position =position_stack(vjust =0.5)) +theme_void()
From the graph, we can see that the gender distribution is relatively balanced, with 53% of patients are female and 47% of patients are male.
plot <-ggplot(data = patients_tble) +geom_histogram(mapping =aes(x = anchor_age, fill = gender), bins =100, alpha =0.8, position ="identity") +labs(title ="Distribution of anchor_age by Gender",x ="anchor_age",y ="Count") +theme_minimal()plotly_plot <-ggplotly(plot)plotly_plot
From the graph, we can see that the distribution of anchor_age is right-skewed, with most patients are between 20 and 25 years old and around 13% of the patients are older than 75 years old.
itemid,label,fluid,category
50801,Alveolar-arterial Gradient,Blood,Blood Gas
50802,Base Excess,Blood,Blood Gas
50803,"Calculated Bicarbonate, Whole Blood",Blood,Blood Gas
50804,Calculated Total CO2,Blood,Blood Gas
50805,Carboxyhemoglobin,Blood,Blood Gas
50806,"Chloride, Whole Blood",Blood,Blood Gas
50808,Free Calcium,Blood,Blood Gas
50809,Glucose,Blood,Blood Gas
50810,"Hematocrit, Calculated",Blood,Blood Gas
We are interested in the lab measurements of creatinine (50912), potassium (50971), sodium (50983), chloride (50902), bicarbonate (50882), hematocrit (51221), white blood cell count (51301), and glucose (50931). Retrieve a subset of labevents.csv.gz that only containing these items for the patients in icustays_tble. Further restrict to the last available measurement (by storetime) before the ICU stay. The final labevents_tble should have one row per ICU stay and columns for each lab measurement.
Hint: Use the Parquet format you generated in Homework 2. For reproducibility, make labevents_pq folder available at the current working directory hw3, for example, by a symbolic link.
Rows: 1622 Columns: 4
── Column specification ────────────────────────────────────────────────────────
Delimiter: ","
chr (3): label, fluid, category
dbl (1): itemid
ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
# Filter the colomns and rows that we are interested inlabevents <- labevents %>%filter(itemid %in% labevent_need)%>%filter(subject_id %in% sub_need)%>%select(subject_id, hadm_id, itemid, charttime, storetime, valuenum)
labevents_tble <-as_tibble(labevents)
# Merge the labevents_tble with icu_stays by subject_id and # make sure the storetime is before intimemerged_data <- labevents %>%merge(icu_stays, by ='subject_id') %>%select(subject_id,itemid, charttime, storetime, intime, outtime, valuenum, stay_id, los) %>%filter(storetime < intime)
Since some multiple values of the same item with the same storetime, we use the one with the latest charttime. If still multiple values of the same item with the same storeitem and charttime, we can use either value or average.
chartevents.csv.gz (https://mimic.mit.edu/docs/iv/modules/icu/chartevents/) contains all the charted data available for a patient. During their ICU stay, the primary repository of a patient’s information is their electronic chart. The itemid variable indicates a single measurement type in the database. The value variable is the value measured for itemid. The first 10 lines of chartevents.csv.gz are
We are interested in the vitals for ICU patients: heart rate (220045), systolic non-invasive blood pressure (220179), diastolic non-invasive blood pressure (220180), body temperature in Fahrenheit (223761), and respiratory rate (220210). Retrieve a subset of chartevents.csv.gz only containing these items for the patients in icustays_tble. Further restrict to the first vital measurement within the ICU stay. The final chartevents_tble should have one row per ICU stay and columns for each vital measurement.
Hint: Use the Parquet format you generated in Homework 2. For reproducibility, make chartevents_pq folder available at the current working directory, for example, by a symbolic link.
# Filter the colomns and rows that we are interested inchartevents_filter <- chartevents %>%filter(itemid %in% vital_need)%>%filter(subject_id %in% sub_need)%>%select(subject_id, stay_id, itemid, charttime, storetime, valuenum)
chartevents_tble <-as_tibble(chartevents_filter)
# Merge the chartevents_tble with icu_stays by subject_id and stay_idmerged_data_ <- chartevents_tble %>%left_join(icu_stays, by =c('subject_id','stay_id'))
# Make sure the charttime is after intime and before outtimemerged_data_ch <- merged_data_ %>%select(subject_id,itemid, charttime, storetime, intime, outtime, valuenum, stay_id) %>%filter(charttime >= intime) %>%filter(charttime <= outtime)
Let us create a tibble mimic_icu_cohort for all ICU stays, where rows are all ICU stays of adults (age at intime >= 18) and columns contain at least following variables
all variables in icustays_tble
all variables in admissions_tble
all variables in patients_tble
the last lab measurements before the ICU stay in labevents_tble
the first vital measurements during the ICU stay in chartevents_tble
The final mimic_icu_cohort should have one row per ICU stay and columns for each variable.
Answer:
Since the admission_tble has been modified in previous question, we need to re-import the data.
Rows: 431231 Columns: 16
── Column specification ────────────────────────────────────────────────────────
Delimiter: ","
chr (8): admission_type, admit_provider_id, admission_location, discharge_l...
dbl (3): subject_id, hadm_id, hospital_expire_flag
dttm (5): admittime, dischtime, deathtime, edregtime, edouttime
ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
mimic_icu_cohort <- icustays_tble %>%left_join(admissions_tble, by =c("subject_id", "hadm_id")) %>%left_join(patients_tble, by =c("subject_id")) %>%left_join(final_labevents_tble, by =c("subject_id", "stay_id")) %>%left_join(final_chartevents_tble, by =c("subject_id", "stay_id"))
From the graph, we can see that the distribution of ICU length of stay of different race are very similar, which are very right-skewed, with most of the patients (75%) stay in the ICU for less than 5 days.
From the graph, we can see that the distribution of ICU length of stay of different insurance are very similar, which are very right-skewed, with most of the patients (75%) stay in the ICU for about 3 days.
From the graph, we can see that the distribution of ICU length of stay of different marital status are very similar, which are very right-skewed, with most of the patients (75%) stay in the ICU for about 3 days.
Min. 1st Qu. Median Mean 3rd Qu. Max.
F 0.001446759 1.052002 1.911053 3.342715 3.649201 110.2323
M 0.001250000 1.108356 1.940694 3.538524 3.763284 103.4990
ggplot(mimic_icu_cohort, aes(x = gender, y = los)) +geom_boxplot(trim =FALSE, outlier.size =0.5) +labs(title ="Distribution of ICU Length of Stay by Gender",x ="Gender",y ="Length of Stay (days)") +theme(axis.text.x =element_text(angle =0, vjust =0.5, hjust=1, size =8))
From the graph, we can see that the distribution of ICU length of stay of different gender are very similar, which are very right-skewed, with most of the patients (75%) stay in the ICU for about 3 days.
plot <-ggplot(mimic_icu_cohort, aes(x = age_intime, y = los)) +geom_hex() +labs(title ="Distribution of ICU Length of Stay by Age",x ="Age",y ="Length of Stay (days)") +theme_minimal()plotly_plot <-ggplotly(plot)plotly_plot
From the graph, we can see that the distribution of ICU length of stay versus age is relatively balanced, with most of the patients (75%) stay in the ICU for less than 30 days, very a few patients stay in the ICU for more than 60 days.
length of ICU stay los vs the last available lab measurements before ICU stay:
plot <-ggplot(mimic_icu_cohort, aes(x = Creatinine, y = los)) +geom_hex() +labs(title ="Distribution of ICU Length of Stay by Creatinine",x ="Creatinine",y ="Length of Stay (days)") +theme_minimal()plotly_plot <-ggplotly(plot)
From the graph we can see that the most patient have creatinine level less than 5, and the length of ICU stay is relatively balanced. Very few patient have creatinine level more than 20, and the length of ICU stay is relatively shorter for those patients.
plot <-ggplot(mimic_icu_cohort, aes(x = Potassium, y = los)) +geom_hex() +labs(title ="Distribution of ICU Length of Stay by Potassium",x ="Potassium",y ="Length of Stay (days)") +theme_minimal()plotly_plot <-ggplotly(plot)
From the graph we can see that for patient have potassium level less than 3.5 or greater than 6.5 always have a relatively shorter length of ICU stay, usually less than 30. However, for patient have potassium level between 3.5 and 6.5, the length of ICU stay is relatively right-skewed, with a long tail of staying more than 60 days.
plot <-ggplot(mimic_icu_cohort, aes(x = Sodium, y = los)) +geom_hex() +labs(title ="Distribution of ICU Length of Stay by Sodium",x ="Sodium",y ="Length of Stay (days)") +theme_minimal()plotly_plot <-ggplotly(plot)
From the graph we can see that for patient have sodium level less than 125 or greater than 150 always have a relatively shorter length of ICU stay, usually less than 30. However, for patient have sodium level between 130 and 150, the length of ICU stay is relatively right-skewed, with a long tail of staying more than 60 days.
plot <-ggplot(mimic_icu_cohort, aes(x = Chloride, y = los)) +geom_hex() +labs(title ="Distribution of ICU Length of Stay by Chloride",x ="Chloride",y ="Length of Stay (days)") +theme_minimal()plotly_plot <-ggplotly(plot)
From the graph we can see that for patient have chloride level less than 90 or greater than 110 always have a relatively shorter length of ICU stay, usually less than 30. However, for patient have chloride level between 90 and 110, the length of ICU stay is relatively right-skewed, with a long tail of staying more than 60 days.
plot <-ggplot(mimic_icu_cohort, aes(x = Bicarbonate, y = los)) +geom_hex() +labs(title ="Distribution of ICU Length of Stay by Bicarbonate",x ="Bicarbonate",y ="Length of Stay (days)") +theme_minimal()plotly_plot <-ggplotly(plot)
From the graph we can see that for patient have bicarbonate level less than 15 or greater than 35 always have a relatively shorter length of ICU stay, usually less than 30. However, for patient have bicarbonate level between 15 and 35, the length of ICU stay is relatively right-skewed, with a long tail of staying more than 60 days.
plot <-ggplot(mimic_icu_cohort, aes(x = Hematocrit, y = los)) +geom_hex() +labs(title ="Distribution of ICU Length of Stay by Hematocrit",x ="Hematocrit",y ="Length of Stay (days)") +theme_minimal()plotly_plot <-ggplotly(plot)
From the graph we can see that for patient have hematocrit level less than 20 or greater than 50 always have a relatively shorter length of ICU stay, usually less than 30. However, for patient have hematocrit level between 20 and 50, the length of ICU stay is relatively right-skewed, with a long tail of staying more than 60 days.
plot <-ggplot(mimic_icu_cohort, aes(x =`White Blood Cells`, y = los)) +geom_hex() +labs(title ="Distribution of ICU Length of Stay by White Blood Cell Count",x ="White Blood Cell Count",y ="Length of Stay (days)") +theme_minimal()plotly_plot <-ggplotly(plot)
From the graph we can see that most patient have white blood cell count less than 50 and have a relatively right-skewed length of ICU stay. However, for patient have white blood cell count between 50 and 100, the length of ICU stay is relatively balanced, with most of the patients stay in the ICU for less than 30 days. Only a few patient have white blood cell count more than 400, and the length of ICU stay is relatively shorter for those patients.
plot <-ggplot(mimic_icu_cohort, aes(x = Glucose, y = los)) +geom_hex() +labs(title ="Distribution of ICU Length of Stay by Glucose",x ="Glucose",y ="Length of Stay (days)") +theme_minimal()plotly_plot <-ggplotly(plot)
From the graph we can see that most patient have glucose level less than 200 and have a relatively right-skewed length of ICU stay. However, for patient have glucose level between 200 and 1500, the length of ICU stay is relatively balanced, with most of the patients stay in the ICU for less than 30 days. Only a few patient have glucose level more than 1500, and the length of ICU stay is relatively shorter for those patients.
Length of ICU stay los vs the average vital measurements within the first hour of ICU stay:
plot <-ggplot(mimic_icu_cohort, aes(x =`Heart Rate`, y = los)) +geom_hex() +labs(title ="Distribution of ICU Length of Stay by Heart Rate",x ="Heart Rate",y ="Length of Stay (days)") +theme_minimal()plotly_plot <-ggplotly(plot)
From the graph, we can see that the distribution of ICU length of stay versus heart rate is relatively right-skewed. Most patient have Heart Rate around 75 and stay for about 3 days, and it’s unusual that some patients have a heart rate that is more than 200.
plot<-ggplot(mimic_icu_cohort, aes(x =`Non Invasive Blood Pressure systolic`, y = los)) +geom_hex() +labs(title ="Distribution of ICU Length of Stay by Systolic Blood Pressure",x ="Systolic Blood Pressure",y ="Length of Stay (days)") +theme_minimal()plotly_plot <-ggplotly(plot)
From the graph, we can see that the distribution of ICU length of stay versus systolic blood pressure is relatively right-skewed. Most patient have Systolic Blood Pressure around 120 and stay for about 3 days, and it’s unusual that some patients have a Systolic Blood Pressure that is more than 12500
plot<-ggplot(mimic_icu_cohort, aes(x =`Non Invasive Blood Pressure diastolic`, y = los)) +geom_hex() +labs(title ="Distribution of ICU Length of Stay by Diastolic Blood Pressure",x ="Diastolic Blood Pressure",y ="Length of Stay (days)") +theme_minimal()plotly_plot <-ggplotly(plot)
From the graph, we can see that the distribution of ICU length of stay versus diastolic blood pressure is relatively right-skewed. Most patient have Diastolic Blood Pressure around 80 and stay for about 3 days, and it’s unusual that some patients have a Diastolic Blood Pressure that is more than 60000.
plot <-ggplot(mimic_icu_cohort, aes(x =`Temperature Fahrenheit`, y = los)) +geom_hex() +labs(title ="Distribution of ICU Length of Stay by Temperature",x ="Temperature",y ="Length of Stay (days)") +theme_minimal()plotly_plot <-ggplotly(plot)
From the graph, we can see that the distribution of ICU length of stay versus temperature is relatively right-skewed Most patient have temperature around 98 and stay for about 3 days, and it’s unusual that some patients have a temperature that is more than 1000.
plot <-ggplot(mimic_icu_cohort, aes(x =`Respiratory Rate`, y = los)) +geom_hex() +labs(title ="Distribution of ICU Length of Stay by Respiratory Rate",x ="Respiratory Rate",y ="Length of Stay (days)") +theme_minimal()plotly_plot <-ggplotly(plot)
From the graph, we can see that the distribution of ICU length of stay versus respiratory rate is relatively right-skewed. Most patient have respiratory rate around 20 and stay for about 3 days, and it’s unusual that some patients have a respiratory rate that is more than 75.
Min. 1st Qu. Median
Cardiac Vascular Intensive Care Unit (CVICU) 0.003518519 1.2134664 1.988229
Coronary Care Unit (CCU) 0.001250000 1.0891319 2.014063
Medical Intensive Care Unit (MICU) 0.002268519 0.9804572 1.829288
Medical/Surgical Intensive Care Unit (MICU/SICU) 0.001446759 1.0151620 1.808368
Neuro Intermediate 0.003460648 1.1225405 2.335116
Neuro Stepdown 0.057002315 0.8905787 1.686250
Neuro Surgical Intensive Care Unit (Neuro SICU) 0.021574074 1.7941406 3.651233
Surgical Intensive Care Unit (SICU) 0.007453704 1.0798958 1.974583
Trauma SICU (TSICU) 0.002175926 1.0318953 1.939421
Mean 3rd Qu. Max.
Cardiac Vascular Intensive Care Unit (CVICU) 3.290002 3.295211 103.49900
Coronary Care Unit (CCU) 3.192368 3.725990 66.58424
Medical Intensive Care Unit (MICU) 3.263419 3.522914 110.23228
Medical/Surgical Intensive Care Unit (MICU/SICU) 3.084816 3.193600 67.87067
Neuro Intermediate 3.420690 4.246418 43.51039
Neuro Stepdown 2.590663 3.163588 29.10352
Neuro Surgical Intensive Care Unit (Neuro SICU) 6.299043 7.743111 59.56295
Surgical Intensive Care Unit (SICU) 3.837683 4.026551 101.72624
Trauma SICU (TSICU) 3.833085 3.983388 99.63845
ggplot(mimic_icu_cohort, aes(x = first_careunit, y = los)) +geom_boxplot(trim =FALSE, outlier.size =0.5) +labs(title ="Distribution of ICU Length of Stay by First Care Unit",x ="First Care Unit",y ="Length of Stay (days)") +theme(axis.text.x =element_text(angle =10, vjust =1, hjust=1, size =6))
From the graph, we can see that the distribution of ICU length of stay of different first care unit are very similar, which are very right-skewed, with most of the patients (75%) stay in the ICU for about 3 days.
Do a similar visualization for the patient with 
Answer:
